JAERI- M

.. RESPONSE DISTRIBUTIONS OF 'LiF AND 'L~F THERMOLUMINESCENCE DOSIMETERS IN LITHIUM BLANKET ASSEMBLIES

I November 1 9 7 6 I I Hiroshi MAEKAWA, Jyo-ichi KUSANO I

and Yasushi SEKI

E l * = j F f i @ f % % Japan Atomic Energy Research Institute

86D600DI

JAERI-M reports, issued irregularly, describe t!ie results of research works carried out

in JAERI. Inquiries about the avai!ability of reports and their reproduction sl~ould be addressed to Division of Teclinica! Information, Japan Atomic Energy Research Institute,

Tokai.mura, Naka-gun, Ibaraki-ken, Japan.

JAERI - M 6811 NEACRP-L-165

Response Distributions of 6 ~ i ~ and 7 ~ i ~ Thermoluminescence

Dosimeters in Lithium Blanket Assemblies

Hiroshi MAEKAWA, Jyo-ichi KUSANO and

Yasushi SEKI

Division of Reactor Engineering, Tokai, JAERI

(Received November 8, 1976)

Measurement of the radiation-heating rate distribution in

the fusion blanket is as important as measurement of the

fission-rate distribution in a fission reactor. To obtain the

information of radiation heating, the response (integral glow

value) distributions in. pseudo-spherical lithium assemblies

with and without a graphite reflector were measured with 6 ~ i ~ - and 'L~ F TLDts. The measured responses are normalized to values per source neutron. Experimental error is about 35 %, and the error in positions of TLD's is about + 3 mm. The

experimental results are compared with those of calculation

using RADHEAT code system and ENDF/B-111 data file.

J A E R I - M 6 8 1 1

I. INTRODUCTION

Fusion energy is converted to thermal energy in the blanket region surrounding the plasma of a fusion reactor. The measure- ment of radiation heating-rate distribution in the blanket is equally important as the measurement of fission-rate distribu- tion in a fission reactor. According to the neutronics calculationc1), heating rate due to gamma-rays in a blanket is about a third of that due to neutrons. At present, there is no direct means for measuring the radiation heating-rate distribu- tion. Some information.on neutron heating can be obtained from our measurement of the fission-rate distribution in the blanket

a assemblies ( 3 ) . Gamma-ray heating-rate distribution in a fission reactor has been measured by G.G. Simons et al. ( 4 )

7 making use of the difference of response of 6 ~ i ~ and LiF thermoluminescence dosimeters (TLD). In a fusion reactor

I blanket where the neutrons with high energy are present, it is

impossible to measure gamma-ray heating selectively by 7 ~ i ~ TLD because 7~i(n,n'a)~ reaction contributes largely to the formation of thermoluminescence (TL).

The reaction of 6~i(n,a)~ and 7~i(n,n'a)~ are the two major reactions contributing to heat deposition in a blanket with

i lithium. When TLD's of 6 ~ i ~ and 'IL~F are employed, the major portion of radiation heating-rate in lithium may be measured. *

, . We have measured the responses of 6 ~ i ~ and 'IL~F TLD1s to the neutron and gamma-ray in pseudo-spherical lithium assemblies with and without a graphite reflector. The results of the measurements are compared with those of calculations. In order to calculate the total response of the TLD, the energy depen- dence of the TLD response to the neutrons and gamma-rays must be known respectively. The energy dependence of the response to gamma-ray are accurately known. As for the energy dependence to the neutrons, the responses of 6 ~ i ~ and 'IL~F TLD1s to various monoenergetic neutrons are measured by Y. Furuta and S. Tanaka, and they were shown to be in good agreement with the calculated

* The unit value is the equivalent 6 0 ~ o gamma-ray exposure in roentgens.

JAERI-M 68 1 1

11. EXPERIMENTAL ASSEMBLIES AND PROCEDURES

1. PRINCIPLE OF MEASUREMENT

The response (integral glow value) of TLD in the unit of X

6 0 ~ o equivalence may be represented by the following equation from the values read from a TLD reader,

Here, R(r); The response of TLD at the position r, in the unit of 6 0 ~ o equivalence per unit source neutron (R/neutron).

9 ; The factor for converting the value indicated in

the reader to 6 0 ~ o equivalence. D(r); The dose of TLD at the position r as indicated by

the reader. N ; Total neutron yield during the measurement.

The conversion factor n may be obtained by irradiating the TLDts in the standard irradiation field of 6 0 ~ o gamma source prior to the experiment.

On the other hand, the response may be represented by the following equation using the calculated neutron and gamma-ray fluxes,

where, Rn: the response-function for neutrons,

% : neutron flux,

RY : the response-function for gamma-rays,

@ Y : gamma-ray flux.

The neutron and gamma-ray fluxes in the experimental assemblies are obtained from the blanket neutronics calculation procedures developed in JAERI for the design of fusion reactors ('I. The response-function for gamma-rays R is calculated

Y

JAERI-M 6 8 1 1

from the mass energy absorption coefficient of TLD. The response-function for neutrons Rn obtained by Y. Furuta and S.

Tanaka * (6) is adopted.

2. EXPERIMENTAL ASSEMBLY

The vertical cross-section of the spherical lithium metal assembly with a graphite reflector (named Li-C Assembly) is shown in Fig. 1. It is constructed by loading stainless-steel

canned lithium metal blocks and graphite blocks in drawers made of stainless-steel, which in turn were loaded on stainless-steel

a lattice matrix. The unit size of the block is about 2 inches cube. The 3 x 3 x 3 units in the center consist only of stain- less-steel drawers and lattice and the region covered by these units are called Void region. The effective radius (meaning the radius of the equivolume sphere) of Void region is 10.0 cm. The effective outer radius of Li-region, where the lithium blocks are loaded, is 34.1 cm, and that of Graphite region with graphite blocks is 55.3 cm.

The spherical lithium assembly without a graphite reflector is called Li Assembly. The details of the experimental assem- blies are given in the Refs. ( 2 ) and (3).

A Cockcroft-Walton type accelerator is used as the neutron generator. The D+ beam of 300 keV is focused on a 10 Ci tritium

a metal target placed at the center of the experimental assemblies. The TLD response is measured in the direction perpendicular to the D+ beam.

3. EXPERIMENTAL PROCEDURES The hot-pressed type TLD-600 and TLD-700 supplied by

Harshaw Chemical Co. are used for the measurement. They are rods 6 mm long with 1 mm square cross-section and their weight is about 15 mg.

Three TLDts (each 6 ~ i ~ (TLD-600) and 7 ~ i ~ (TLD-700)) are inserted in a small envelope of polyethylene and fixed by cellophane tape in the space for experimental hole in the stainless-steel drawers for each experiment point. The detector

J A E R I - M 6811

positions are in the direction perpendicular to the incident D+

beam on the same holizontal plane as the target as shown in Fig. 1. An a-monitor is placed in the beam extension tube, at 179

degrees to the D+ beam and about 107 cm from the target, to count the a-particles emitted by D-T reaction and to obtain the

absolute neutron yield.

After irradiation, the TLD's were removed immediately

from the assembly. They were left for 24 hours in room tempera-

ture, and the irradiation dose were measured by the TLD reader.

Prior to the irradiation, the TLD's are calibrated by 6 0 ~ o

source with strength 7.2 mCi, in order to obtain exposure dose.

The conversion factor q is calculated from the data. The TLD's

having the dose within t 3 % of the average were selected for

use in the experiment. 0

111. CALCULATIONAL PROCEDURE

The neutronics calculations are based on RADHEAT(~~, the

code system for calculating radiation heating-rate. The 42

group neutron transport cross-sections of up to P5 Legendre

terms are obtained by SUPERTOG (81 from ENDF/B-I11 data file

The 21 group gamma-ray transport cross-sections are calculated

by GAMLEG-JR(~). The energy structures of neutron and gamma-

ray cross-sections are shown in Tables I and 11, respective'ly.

The gamma-ray production cross-sections are calculated by POPOP4

from POPOP4 library (101

The calculational models for Li and Li-C Assemblies are

shown in Fig. 2. Neutron and gamma-ray fluxes are calculated

by one-dimensional transport code ANISN employing P5 - S8 approximation. The group-wise responses of TLD's for neutrons

and gamma-rays are calculated from the response-functions Kn

and R respectively. The calculational response of TLD is the Y'

sum total of them. There is a little difference in the gamma-.

ray response functions for 6 ~ i ~ and 7 ~ i ~ TLD's, but the value

for the natural lithium fluoride TLD is used in the present

calculation.

IV. RESULTS AND DISCUSSIONS

The measured responses of 6 ~ i ~ and 7 ~ i ~ TLD's in Li and Li-C Assemblies are compared with calculated ones in Figs. 3 and 4, respectively. Since both measured and calculated values are normalized to values per one source neutron, direct comparison of absolute values are possible. Experimental error amounts to about 35 %, which is the sum of following errors;

the deviation of TLD elements 5 %,

the statistical reading error 10 %,

the total neutron yield error 5 % ,

the measurement error of conversion factor 15 % .

The error in the position of TLD is about + 3 mm. Considering the experimental error and the accuracy of

gamma-ray data using the calculation, the measured and calcu- lated values in the Li Assembly are in fairly good agreement for both 6 ~ i ~ and 'L~F for all except the first measured point (at 8.3 cm). But in Li-C Assemblies, while the measured and calculated values for 7 ~ i ~ are in fairly good agreement, a large difference between them is observed for 6 ~ i ~ .

Each response for neutrons and gamma-rays in the two assemblies is calculated for 6 ~ i ~ and 'L~F TLD's. They are shown in Figs. 5 and 6. The calculated curves in Figs. J and

0 4 are the sum of neutron and gamma-ray responses in Figs. 5 and 6, respectively. The magnitude of the response to gamma- rays is of comparable order to that of the response to neutrons in the lithium region of Li-Assembly as shown in Fig. 5. It may also be observed from the figure that the response of 'L~F TLD to neutrons cannot be neglected when 14 MeV neutrons are

present. The response of 7 ~ i ~ TLD in Li-C Assembly is not much different from that in Li-Assembly. The 'response of 6 ~ i ~ TLD to neutrons is far greater than that to gamma-rays in the graphite region of Li-C Assembly. Therefore, 'the large differ- ence observed between-the calculations and experiment about

6 the LiF response in the graphite region may be attributed to the calculational error in low energy neutron flux, which contri

J A E R I - M 6 8 1 1

bute to 6~i(n,a)~ reaction.

Similar results have been obtained from the absolute

fission-rate distribution in Li-C Assembly (11), and the major

cause of the discrepancy between measurement and calculation

is identified as the inaccurate treatment of the anisotropy

of nonelastic neutrons in processing multigroup cross-sections

from nuclear data c.12 1 Large discrepancy observed near the center of the assembly

are explained by treating the shape of neutron source as a small

sphere in the calculation, by the fluctuation of the beam spot,

and by the gamma-rays caused by the accelerated deuteron hitting

the target which have been neglected in the calculation.

The following problems in the experimental procedures

resulted in large errors:

(1) The error in the caxibration of TLD elements, i.e. in

the conversion factor , is large. There is 5 % error in

the strength of 6 0 ~ o gamma-ray source itself and additional

error is caused by the back-scattered gamma-rays from the

floor and wall of the calibration room and also by the

scattered gamma-rays from the structural materials for

supporting the TLD elements and source.

(2) The reading error for low dose-rate by the TLD reader

is large.

(3) Although TLD elements with least fluctuation is selected

prior to the experiment, the fluctuation is still unaccept-

ably large.

The following problems in the calculational procedures

are pointed out:

(4) The calculated gamma-ray flux may have large error owing

to the insufficient gamna-ray production data in the

POPOP4 Library.

(5) The neutron cross-section processing code is inadequate

for the analysis of fusion blanket neutronics because the

anisotropy of nonelastic neutrons is not properly accounted

for as already described in Ref. (12).

JAERI-M 6811

(6) One-dimensional sphere models used in the analysis does not reflect the details of the experimental assemblies.

(7) The stainless-steel structural materials near the target is accounted for in the calculations by including them into the homogenized atomic density of the region from 0.5 cm to 3.3 cm. Since the gamma-ray is produced mostly near the target, the search for the best model is required for the better description.

In spite of all the uncertainties accompanying the present experiment and analysis, the measurement of the distribution of

7

a 6 ~ i ~ and LiF TLD response in the fusion blanket mock-up assem- blies is very promising as an effective means for obtaining the

information on radiation heating-rate distribution.

ACKNOWLEDGMENT

The authors would like to acknowledge Drs. J. Hirota and H. Kuroi for their encouragement. They are also indebted to Mr. S. Tanaka and Dr. Y. Furuta for helpful discussions.

REFERENCE

SEKI, Y. : "Evaluation of Shielding Design of Super

Conducting Megnet (I)", JAERI-M 6046 (1975) (in Japanese)

MAEKAWA, H. et al. : Nucl. Sci. Eng. - 57, 335-340 (1975)

MAEKAWA, H., SEKI, Y. : "Absolute Eission-Rate

Distribution in Lithium and Hybrid Fusion Blanket

Assemblies, I Experimental Method and Results", To be

published in J. Nucl. Sci. Technol.

SIMONS, G.G. et al. : Nucl. Sci. Eng. - 53, 162-175 (1974)

FURUTA, Y., 'IANAKA, S. : Nucl. Inst. Meth. - 104, 365-374

(1972)

FURUTA, Y., TANAKA, S. : "The Relation Between Light

Conversion Efficiency and Stopping Power of Charged 0

Particles in Thermolumine cence Dosimeter", International

Symp. on Radiation Physic;, Calculta (1974)

MIYASAKA, S. et al. : "Code System for the Radiation-

Heating Analysis of a Nuclear Reactor RADHEAT", JAERI-M

5794 (1974)

WRIGHT, R.Q. et al. : "SUPERTOG: A Program to Generate

Fine Group Constant and Pn Scattering Matrices from ENDF/B9',

ORNL-TM-2679 (1969)

FORD, W.E., III., WALLACE, D.H. : "POPOP4, A Code for

Converting Gamma-Ray Spectra to Secondary Gamma-Ray '

Froduction Cross-Section", CTC-12 (1969)

FORD, W.E., III., "The POPOF4 Library of Neutron Induced

Secondary Gamma-Ray Yield and Cross Section Data", CTC-42 a

(1970)

SEKI, Y., MAEKAWA, H. : J, Nucl. Sci. Technol. - 13, 272-

275 (1976)

SEKI, Y., MAEKAWA, H. : "Absolute Fission-Rate Distribu-

tions in Lithium and Hybrid Fusion Blanket Assemblies, 11.

Analysis and Evaluation", To be published J. Nucl. SCL.

Technol

JAERI-M 6811

Table 1 Neutron Energy Structure

Group

1 2 3 '3

Energy Lirnik Mid - Point Energy

Table11 Gamma-Ray Energy Group Structure

GrOLp Energy Limits (MeV) Mid- Point Energy (W)

Fig. 1 V e r t i c a l Cross Sect ion of t h e Spher ica l Lithium Metal

Assembly wi th a Graphite Ref l ec to r

JAERI-M 6811

I s o t r o p i c Neutron

0.5

L i Assembly

I s o t r o p i c Neutron Source

Graphite L a t t i c e R

0 3.3 10.0 34.1 55.3 70.0 0.5

Li-C Assembly

Fig.2 Ca lcu la t iona l Model o f t h e Spher i ca l

Lithium Metal Assembly

JAERI-M 6811

R a d i u s ( c m )

Fig. 3 Response Distributions of 6 ~ i ~ and T L ~ F TLD's in Li Assembly

L - p Void -4- L i t h i u m -7-2- G r a p h i t e 7 - - I

I I I

I I - -16 I I I I I I I I I

R a d i u s ( c m )

Fig. 4 Response Distributions of 6 ~ i ~ and Y L ~ F TLD's in Li-C Assembly

0 -C,

F ~ o i c i - i = I

Lithium ------I I

Rad ius (cm)

Fig. 5 Calculated Response Curves of LiF TLD for Neutrons and Gamma-rays

in L i hs.embly

- lo--@ . C P Li - C Assembly 6 5 C ---- 1. Response of L iF for Neutrons 4- 3 n -.- 7 . CI ,.. h Response of L IF for Neutrons

-3

.' " 0 10 20 30 40 5 0 60

Radius ( c m )

Fig. 6 Calculated Response Curves of LiF TLD for Neutrons Ad Gamma-rays i n L i -C Assembly